ResearchBenchmarksReinforcement Learning on antmaze large-play offline-to-onlineReinforcement Learning on antmaze large-play offline-to-onlineFollow93ScoreRLPD54.5264.5174.584.49May 14, 2026Evaluation resultsMethodMethodLinksScoreRLPDEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0593CQLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0587.3CPQLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0585.4Cal-QLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0575PEXEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0556